Browsing transcripts

The primary way to browse transcripts in APLS is via the transcripts page: https://apls.pitt.edu/labbcat/transcripts.

The transcripts page lists all the transcripts in APLS that are available for access, but it also offers useful tools for filtering and exporting transcripts. On this documentation page, we cover the functionality and layout of the transcripts page.

On this page
  1. What you can do on the transcripts page
    1. Layout
  2. Transcript list
  3. Filtering transcripts
    1. Clearing filters
  4. Exporting and searching transcripts
    1. Export Media and Export Original
    2. Export Attributes
    3. Export Formatted
    4. Layered Search
  5. Transcript attributes pages

What you can do on the transcripts page

The transcripts page allows you to…

Layout

Transcript list

The transcript list displays all transcripts in APLS that match your filter criteria.

The transcript list in the screengrab above shows all transcripts in APLS that are of the interview type, tagged as Cranberry Township neighborhood, and have a duration between 300 and 500 seconds.

The columns in the transcript list, from left to right, are:

  • A checkbox that can be toggled on to have the transcript included in Export menu options.
  • The name of the transcript.
    • Clicking on the transcript name will open the transcript page for that transcript.
  • The content type of the transcript.
  • The neighborhood of the transcript.
  • The duration of the transcript in seconds.
  • A clickable attributes icon that will open that transcript’s attributes page.

Go to the transcripts page and click CB05pairs.eaf to view that transcript’s transcript page.

You can read more about transcript attributes in the field guide.

Filtering transcripts

The filter menu at the top of the transcripts page lets you find transcripts that match certain criteria.

Underneath the Transcripts heading is the match count, which shows the number of transcripts that currently fit the criteria of your filters. In the screengrab above, no filters have been applied so the match count displays 218,1 which is the total number of transcripts in APLS.

The four filter fields correspond to the columns in the transcript list directly below the filter fields:

  • The “Transcript name” text field filters transcripts by name and supports regular expressions.

  • The “Transcript type” multi-choice menu filters transcripts by their type of content.

    Because transcript type is determined by the content of the transcript, it makes it easy for researchers to filter for certain types of speech. To view all transcripts with higher attention to speech, go to the transcripts page and select reading and pairs from the Transcript type drop-down list.

  • The “Transcript neighborhood” multi-choice menu filters transcripts according to the Pittsburgh neighborhood where the participant was recruited from.
  • The “Duration (sec)” text fields filter transcripts by their duration in seconds.
    • To view transcripts that are…
      • at least X seconds long: enter X in the From box (leave To blank)
      • at most Y seconds long: enter Y in the To box (leave From blank)
      • between X and Y seconds long (inclusive): enter X in the From box and Y in the To box

    Go to the transcripts page and enter 50“into the From text field and leave the To text field blank. This will show all transcripts that are at least 50 seconds in duration.

    In the same way, you can leave the From text field blank and enter 300 in the To text field to show all transcripts that are 300 seconds or less in duration.

Clearing filters

The “delete” button () allows you to clear all currently specified filters.

Exporting and searching transcripts

The export menu allows you to download transcripts in a variety of formats and perform searches on transcripts. The export menu is located below the filter menu.

If no transcripts are selected, then these options will export all transcripts that match your current filter criteria.

Export Media and Export Original

The most straightforward export options are Export Media and Export Original.

  • Export Media will download the audio for the selected transcripts as .wav files (packaged in a .zip file if more than one transcript is selected)
  • Export Original will download the original ELAN transcripts for the selected transcripts as .eaf files (packaged in a .zip file if more than one transcript is selected)

Export Attributes

The Export Attributes option allows you to download the metadata attributes for the selected transcripts. Clicking the Export Attributes button will bring up a multi-select menu that expands when you hover it, as seen below.

Descriptions of the different transcript attributes can be found in the field guide.

After selecting the attribute data you would like to export, click the Export Attributes button again to download the attributes for all selected transcripts as a single .csv file.

Export Formatted

The Export Formatted option allows you to download layers from transcripts in a variety of file types.

Clicking the checkbox next to a layer in this menu will select that layer to be included in the downloaded files. You can use Shift+Click to select multiple checkboxes at once in this menu.

The attribute typology and field guides provide descriptions of transcript attributes and participant attributes. The layer field guide documentation page provides descriptions of all layers in APLS.

Below the layers selection menu, there is a drop-down menu that allows you to select the file format for the transcript export download.

The following formatted file types are available with Export Formatted:

  • LaTeX Document (.tex)
  • Praat TextGrid (.TextGrid)
  • PDF Document (.pdf)
  • CLAN CHAT transcript (.cha)
  • ELAN EAF Transcript (.eaf)
  • EMU-SDMS Bundle (.json)
  • Transcriber transcript (.trs)
  • Comma Separated Values (.csv)
  • WebVTT subtitles (.vtt)
  • SALT transcript (.slt)
  • Plain Text Document (.txt)

After selecting your desired layers and file format type, click the Export Formatted button again to export the transcripts.

If more than one transcript is selected, the exported files will be packaged in a single .zip file.

The Layered Search option will open the search page with the selected transcripts in the Transcripts search filter field. This allows you to use any of the search capabilities described on the Searching the corpus documentation page with the selected transcripts.

  1. Go to the transcripts page.
  2. Click the checkbox next to CB01interview1.eaf and Shift+Click the checkbox next to CB01reading2.eaf to select all CB01 transcripts.
  3. Click Layered Search to open the search page.
  4. Enter steel into the Regular expression text field in the orthography section and click Search.
  5. Click Display results to view all utterances of the word steel in the CB01 transcripts.

Transcript attributes pages

Clicking the attributes icon for a transcript will open that transcript’s attributes page. This page includes more information about the transcript than what is displayed on the main transcripts page, as well as a link to display information about the participants in the transcript.

The Participants hyperlink will open the Participants page for the “Main speaker” participants present in the transcript.

  1. Go to the transcripts page
  2. Click the attributes icon for CB01reading1.eaf to view the transcript’s attributes page.
  3. Click Participants to view CB01 on the Participants page.

A detailed description of the different transcript attributes can be found in the field guide.

  1. For multi-choice filter options, selecting none of the options is the same as selecting all. ↩